Vehicle Surround Display Using Motion-Based Blind Area Rendering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current vehicle surround view systems have blind spots, particularly at ground level, which are not captured by cameras mounted on the vehicle, leading to visually unappealing and potentially irritating placeholders in the display for passengers.
Innovation Solution
A method and device that synthesize and update blind area images in real-time using motion data to predict and render the vehicle's surroundings outside the camera's field of view, eliminating the need for placeholders and optimizing storage resources by using a recursive image processing approach.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If vehicle cameras are mounted on the vehicle front, rear, and sides to detect multiple fields of view, then the coverage of vehicle surroundings is improved, but blind areas remain outside all fields of view that cannot be captured
Solution Approach 1:
The system performs preliminary action by capturing a reference image before vehicle motion occurs, then uses this reference image to synthesize blind area images through warping transformations based on detected vehicle motion, thereby proactively filling information gaps before they become visible defects in the display
Solution Approach 2:
The system creates a copy of the reference image and applies warping transformations to generate synthesized blind area images, using the reference image as a template to reproduce visual information for areas not directly captured by cameras
2Device complexity
If placeholders are used to occupy blind area regions in the display, then the display structure is simplified, but the visual appearance becomes unappealing and irritating for passengers
Solution Approach 1:
Instead of using simple placeholders, the system copies and transforms reference image data to create realistic synthesized images that visually match the surrounding environment, thereby maintaining visual appeal while avoiding complex additional hardware
3Measurement precision
If multiple blind area images are stored for different time points, then the accuracy of temporal reconstruction is improved, but storage resources are significantly consumed
Solution Approach 1:
The system performs preliminary action by capturing a single reference image before motion occurs, then uses this pre-captured image combined with real-time motion detection to synthesize blind area images for multiple time points, thereby achieving temporal reconstruction accuracy while minimizing storage requirements
Solution Approach 2:
The single reference image serves multiple functions: it is used to generate synthesized blind area images for multiple different time points through warping transformations, thereby achieving multi-time-point reconstruction accuracy while consuming minimal storage resources
4Loss of information
If blind area images are synthesized in real-time during vehicle operation, then the display comprehensiveness is improved, but computational resources and processing time are increased
Solution Approach 1:
The system performs preliminary action by capturing a reference image before vehicle motion, then uses pre-detected motion data to warp and synthesize blind area images, thereby reducing real-time computational burden while maintaining display comprehensiveness
Solution Approach 2:
The system uses feedback from vehicle motion detection (odometry data, IMU sensors) to dynamically adjust the warping parameters of blind area image synthesis, thereby optimizing computational resource usage based on actual vehicle movement while maintaining comprehensive display coverage
Data Source
AI summary
A device and a method for displaying vehicle surroundings in a vehicle during an instantaneous second time point are provided. The method includes providing a first blind area image that contains an image synthesis of a blind area of the vehicle surroundings at a first time point preceding the second time point, arranging each first blind area pixel of the blind area in a new position estimated for the second time point, determining whether the new position of each first blind area pixel at the second time point still lies within the blind area, and producing a respective second blind area pixel for the second time point by synthesizing each first blind area pixel on the basis of motion data of the vehicle if the new position is determined to lie within the blind area.